ChatGPT, is a chatbot model developed by OpenAI.
The model has been trained on a diverse range of materials, including website content writing, articles, and even books -- allowing it to generate human-like texts.
So far, the bot has completed and passed several assignments, quizzes, and tests, using GPT-3 (Generative Pre-trained Transformer 3), an autoregressive language model architecture.
Initially launched on November 30, 2022, ChatGPT quickly gained attention for its detailed responses and articulate answers across many domains of knowledge.
However, the bot's factual accuracy in its general texts, remains a massive drawback.
Here is a list of the tests the AI bot passed
Essay writing
ChatGPT's prompt essay writing on almost all topics has been a serious concern for teachers.
They say assignments or giving students an eassay as homework days are over. However, the educators are worried about plagiarism in the texts.
In December last year, Bloomberg podcaster Matthew S Schwartz noted that he had fed a law school essay prompt into ChatGPT and it had "responded instantly with a solid response."
However, a philosophy professor at Furman University had a different story to tell.
He caught a student who submitted an AI-generated essay, and said the eassy had “well written misinformation.”
"Word by word it was a well-written essay," the professor told Insider.
However, he later noticed a part in the essay about another philosopher, that made no sense.
US medical exam
Researchers had put ChatGPT through a three-part United States Medical Licensing Exam, and reported that the bot “performed at or near the passing threshold for all three exams without any specialized training or reinforcement. Additionally, ChatGPT demonstrated a high level of concordance and insight in its explanations.”
Their findings were published in a paper in 2022. The research is still under peer review.
However, the abstract of the paper states that the bot may have "the potential" to assist with medical education. The researchers also concluded that ChatGPT can even perform clinical decision-making.
Law course exams
The AI bot passed four law course exams at the University of Minnesota, US.
The professors blindly graded over 95 multiple-choice questions and 12 essay questions that ChatGPT answered.
ChatGPT secured low but passing grades in the courses, which is equivalent to a C+.
Microbiology test
Alex Berezow, a science journalist and executive editor of Big Think, tested ChatGPT with a microbiology quiz.
The quiz had 10 questions. Berezow noted that the questions were designed for college students.
ChatGPT "blew it away," Berezow wrote, as the AI bot responded to the questions correctly.
